# Using FACTS devices and energy storage systems for voltage flicker mitigation in electric arc furnaces.

IntroductionAn electrical arc furnace (EAF) changes the electrical energy into thermal energy by electric arc in melting the raw materials in the furnace. During the arc furnace operation, the random property of arc melting process and the control system are the main reasons of the electrical and thermal dynamics. That will cause serious power quality problems to the supply system [1].

The fundamental component of the current drawn by an EAF produces fluctuations of the voltage in the nearby distribution system. These fluctuations are the reasons of the phenomenon known as flicker. The voltage changes as much as 0.3~1% with frequencies between 2 and 8 Hz [2, 3].

Building new lines, installing new and bigger transformers, or moving the point of common coupling to a higher voltage level are the traditional methods to deal with problem of poor power quality in distribution system. These methods are expensive and time-consuming. Installing the compensation equipment in the immediate vicinity is a straightforward and cost-effective way of dealing such problem [4].

An equally rapid compensating device is required to remedy and prevent the spreading of the power quality problem caused by EAF. Currently, the most widely used method for flicker compensation is the connection of shunt static VAR compensators based on thyristor-controlled reactors (TCR's). A TCR consists of a reactance connected in series with a pair of thyristors with a fixed value parallel-connected capacitor [3]. These methods, used with conjunction with fixed passive filters, have been successful in correcting the power factor and compensating the harmonics.

Even though these methods have success in solving the flicker problem by reactive power compensation, they are unable to supply any portion of the fluctuating real power drawn by the furnace. Development of high power electronics offers Flexible AC Transmission system (FACTS) several significant advantages, including the ability of passing real power between ac and dc terminals [5].

Integrating an energy storage systems (ESS), such as battery energy storage system (BESS) or superconducting magnetic energy storage (SMES), into a FACTS device can lead to improved controller flexibility by providing dynamic decentralized active power capabilities. Combined FACTS/ESS not only can improve power flow control, oscillation damping, and voltage control, but also improve the power quality of the transmission and distribution systems, including mitigation of the voltage flickers caused by EAF [6].

There are controversial arguments about the effectiveness of active power compensation to solve voltage flicker problems. The reasons given are that the active injection is not effective given that most systems have an X/R ratio over 10 by looking at the transformers and transmission lines. Our study is able to resolve this issue and clearly show that the active power injection is also useful to solve voltage problems. The key here is the calculation of X/R ratio which should be the Thevenin impedance of the system seen at the point of common coupling (PCC). The system X/R ratio turns out to be much lower than 10, for example, X / R [approximately equal to] 3 in this study system.

In this paper, voltage flicker problem caused by EAF in a 25 bus sample power system will be studied. The effect of the random active power drawn by EAF will be discussed in detail in relation to the role of X/R ratio. Using the models in PSS/E, the authors studied the EAF voltage flicker mitigation by FACTS/ESS. The comparison of the mitigation effects by different FACTS/ESS has been shown. The effectiveness of FACTS with and without ESS is compared.

Arc Furnace Model Using PSS/E

An accurate three-phase arc furnace model is needed for the purpose of harmonic analysis and flicker compensation. Since the arc melting process is a dynamic stochastic process, it is difficult to make a precise deterministic model for an arc furnace load. The factors that affect the arc furnace operation are the melting or refining materials, the melting stage, the electrode position, the electrode arm control scheme, the supply system voltage and impedance [1].

Many complex methods were proposed to more precisely represent EAF characteristics and study its impacts on power systems. These include nonlinear resistance model, current source models, voltage source models, nonlinear time varying voltage source model, nonlinear time varying resistance models, frequency domain models, and power balance models, etc, [2,7,8].

PSS/E power system simulation software we chose to use for this study has the ability to deal with a large scale power system. It contains a very large power equipment model library. However, there is no EAF model in it. The proposed EAF model contains a resistor and a reactor in a random operation mode to display the dynamic characteristics of EAF. Because the focus here is to study the voltage flicker problem solutions by FACTS/ESS, a simple arc model will serve the purpose as long as the worst case voltage fluctuation frequency and magnitude are reflected in the model. The proposed arc model can generate a variation of voltage at about 0.3 ~ 1% at a frequency around 5 Hz [2, 3]. It should be mentioned here that harmonic problem is not considered.

The general scheme of EAF and FACTS/ESS is shown in Figure1. In this research, the EAF is about 40 MVA containing 34 MW active power and 25 MVAR reactive power at the normal bus voltage.

[FIGURE 1 OMITTED]

Figure 2 is the 25 bus sample system. The scheme of EAF and FACTS/ESS has the same configuration as shown in Figure 1.

In Figure 2, bus 153 (161 kV) is the PCC bus. The EAF and the FACTS/ESS are connected to bus 1531 (13.8 kV) which is referred to as the EAF bus in this paper. EAF is characterized by rapid changes in absorbing power that occur especially in the initial stage of melting, during which the critical condition of the emerging arc may become a short circuit. Figure 3 shows the characteristic of rapid active and reactive power drawn by EAF. Of course the actual real and reactive power drawn by the EAF does not always change as rapidly as shown in Figure 3. The worst situation for FACTS/ESS to handle is shown in Figure 3. This situation will present the maximum challenge for the compensation equipments. Figure 4 shows the voltage flicker on the PCC bus and EAF bus. The voltage dip is about 1% at frequency around 5 Hz.

[FIGURE 2 OMITTED]

[FIGURE 3 OMITTED]

[FIGURE 4 OMITTED]

Flicker Mitigation Using FACTS/ESS

FACTS/ESS controllers are power electronic based devices that can inject both real and reactive power to not only enhance transmission system performance but also to solve power quality problems. FACTS controllers can be connected to the system in series, in parallel or in combination since they can utilize or redirect the available power and energy from the ac system. Without energy storage, they are limited in the degree of freedom in which they can help the power grid capability.

In this study, two of the available energy storage technologies, battery energy storage system (BESS) and superconducting magnetic energy storage (SMES), are added to a STATCOM to improve the control actions of FACTS.

Figure 5 shows the general configuration of FACTS/ESS while Figure 6 is an example of the FACTS with the super conducting magnetic energy storage device (SMES) [9].

In this study, the capacity of the FACTS/ESS is chosen to meet the demand of the EAF. As the EAF capacity in this study is 40 MVA, the FACTS/ESS in the simulation is 50 MVA with 20 MW active capacity. In this study, the EAF voltage flicker mitigation is achieved by STATCOM, FACTS/BESS and FACTS/SMES. The following are some simulation results to show the effects of the above controllers.

[FIGURE 5 OMITTED]

[FIGURE 6 OMITTED]

Without loss of generality, the energy stored in ESS is limited. If the FACTS/ESS does not restore its active power from the system, the ESS will run out of its active energy (active power) and will behave like a STATCOM, which can only output reactive power. However, the STATCOM can improve the EAF bus voltage as well as the FACTS/ESS does, because the reactive power control scheme of FACTS/ESS is local bus (EAF bus) voltage control. But it cannot improve the voltage of PCC bus as well as the FACTS/ESS because of the lack of active power output. The reasons are discussed in the X/R ratio discussion section. Figure 7 to Figure 9 show the operation mode of the FACTS/SMES when it does not absorb active power from system.

In Figure 7, the energy stored in SMES is depleted within 0.5 seconds if active power is not absorbed from the system. Once this occurs, the reactive power output of FACTS/SMES will rise to control the EAF bus voltage to the desired level. At this time, the FACTS/SMES acts like a STATCOM.

[FIGURE 7 OMITTED]

As the active power output of FACTS/SMES decreases and diminishes, the system will transfer more active power and reactive power to the EAF, which is shown in Figure 8. What is interesting is that the transferring reactive power improves as well. That is because of the voltage drop on the Rn which will be explained in the X/R ration discussion section. In Figure 9, it is evident that even though the EAF bus voltage can be improved to the desired level, the PCC bus voltage cannot be improved to the original reference value when FACTS/SMES only outputs reactive power.

[FIGURE 8 OMITTED]

[FIGURE 9 OMITTED]

In this study, the FACTS/ESS controller task is to absorb active power during the period when EAF is taking less active power from the system. In this operation mode, the FACTS/ESS can work in a continuous mode to supply active power during the whole EAF operation period. The absorption of active power of EAF can cause voltage drop at the PCC and EAF bus, but these voltage drops are relatively small. Again, the X/R ratio plays an important role.

In this continuous operation mode, the reactive power output of FACTS/ESS will not be zero even when EAF is out, as it attempts to compensate the voltage drop caused by the active power draw by ESS as shown in Figure 10.

[FIGURE 10 OMITTED]

In practice, the active and reactive power change of EAF is not as steep as shown in this paper. EAF have a period of rising time, which means the slope of the active and reactive power must not be as sharp as shown in the Figure 3. The spikes in Figure 9 through Figure 17 will not be so high. In Figure 11, the effects of voltage flicker mitigation at the PCC bus and EAF bus are the same.

[FIGURE 11 OMITTED]

In Figure 12 and Figure 13, it is clear that the mitigation effects of BESS and SMES are the same which is better than that of STATCOM. But all of these (STATCOM, BESS and SMES) mitigate voltage flickers caused by EAF effectively. At the EAF bus, the three solutions are equitable.

[FIGURE 12 OMITTED]

In Figure 13, the PCC bus voltage is higher when compensated by STATCOM than by FACTS/ESS. The reason is that the FACTS/ESS absorbing active power causes a voltage drop on PCC bus. Because the control scheme of the FACTS/ESS is to control the EAF bus voltage to the desired level, the PCC bus voltage cannot be improved as much as the EAF bus. This will be explained in section 4.

[FIGURE 13 OMITTED]

Effect of X/R ratio

As mentioned above, the voltage effect of active power drawn by EAF is reflected by the real part of system impedance (Thevenin Impedance) Rn. A very common mistake is to assume that in the power system, the X/R ratio is large enough to omit the voltage drop caused by the active power of EAF. An incorrect approach to calculating

X/R ratio is by considering only the upper transformer or branch and one will certainly arrive at an X/R over 10. The correct approach should be looking at all the parameters including the load. In other words, the R and X are not proportional to the impedance of the upper transformers and lines, they should be the Thevenin impedance seen from the PCC bus of the entire system.

The following example will show how the value of X/R ratio at PCC can be roughly estimated. At the PCC bus, the whole power system can be seen as a power source connected with an active load and a reactive load as shown in Fig.14 (a). The Thevenin Equivalent circuit is shown in Fig.14 (c).

[FIGURE 14 OMITTED]

In the studied system, according to the simulation results, the total active power ([P.sub.L[SIGMA]]) is 3200 MW while the total reactive power ([Q.sub.L[SIGMA]) is 1950 MVAR; [V.sub.LLbase] is 21.6 kV whereas [S.sub.base] is 100 MVA. The impedance of one generator is [Z.sub.s1] = 0.01 + j 0.3 pu. Since [Z.sub.base] = [V.sup.2.sub.LLbase] / [S.sub.base = 4.666 [OMEGA], then [Z.sub.s1] is 0.04666 + j 1.3998?. There are six generators in the studied system. Assuming they are connected in parallel to supply power to the load shown in Figure 14 (a), then the whole system impedance [Z.sub.s] is 1/6 of that of one generator, accordingly Zs will be equal to 0.0079 + j 0.2333 [OMEGA]. The calculated total load impedance is:

[R.sub.1] = [V.sup.2.sub.LLbase] / [P.sub.L] = 0.1458 [OMEGA] while [X.sub.L] = [V.sup.2.sub.LLbase] / [Q.sub.L] = 0.2393 [OMEGA]. As shown in figure 14 (b-c), the values of Rn and Xn can be computed as [R.sub.n] + j[X.sub.n] = 0.0589 + j0.0716 [OMEGA]

with ([X.sub.n] = 1.22. In this example, the values of [X.sub.n] and [R.sub.n] are not precise but enough to demonstrate that the system X/R ratio is affected significantly by the load. After adding impedances of adjacent transformer and transmission line, X/R is still not large enough to omit R at EAF bus. In our case, the system Thevenin Impedance ([R.sub.n]+j [X.sub.n]) seen at PCC bus is 0.00782 +j 0.02389 pu, which was calculated precisely in PSS/E. The [X.sub.n]/[R.sub.n] ratio is about 3. In this case, the active load can cause obvious voltage drop as seen in the above simulation results. In other words, the active power of FACTS/ESS can play an obvious role in voltage compensation as shown in the above study results.

A detailed analysis of the X/R ratio on voltage impact is explained as: the equivalent impedance ([X.sub.n], [R.sub.n]) shown in Figure 15 is the Thevenin equivalent of the whole sample power system looking at PCC bus. [U.sub.o] is the source voltage of the Thevenin equivalent circuit. The X/R ratio is not that of the upper transmission line or upper transformer but that of the Thevenin equivalent impedance ([X.sub.n], [R.sub.n]).

[FIGURE 15 OMITTED]

In Figure 15, the voltage drop of EAF is U ??which contains a real part p U ??and a reactive part q U ??shown in the following equations:

[MATHEMATICAL EXPRESSION NOT REPRODUBLE IN ASCII.] (1)

[MATHEMATICAL EXPRESSION NOT REPRODUBLE IN ASCII.] (2)

[MATHEMATICAL EXPRESSION NOT REPRODUBLE IN ASCII.] (3)

[I.sub.np] and [I.sub.nq] are given as:

[MATHEMATICAL EXPRESSION NOT REPRODUBLE IN ASCII.] (4)

[MATHEMATICAL EXPRESSION NOT REPRODUBLE IN ASCII.] (5)

In this case, the active power and reactive power of EAF is 34 MW and 25 MVAR respectively. Figure 16 shows the voltage drop by 34 MW active load, 25 MVAR reactive load and 40 MVA (34 MW + j25 MVAR) combined load. The calculated ratio of voltage drop caused by 34 MW active load and 25 MVAR reactive load is about 0.445, which is given by equation 6.

[MATHEMATICAL EXPRESSION NOT REPRODUBLE IN ASCII.] (6)

In which cos [phi] = 0.8 and [R.sub.n] = 0.3273 [X.sub.n].

[FIGURE 16 OMITTED]

The simulation results ensure the previous X/R ratio analysis. It is important to pay attention to the role of active power compensation when the voltage flicker caused by EAF is being studied. FACTS with ESS could be a better device to mitigate the voltage flickers caused by EAF.

Figure 17 and Figure 18 show the comparison of voltage flicker mitigation effect by FACTS/BESS with and without active power output. It is evident that for the EAF bus, the effects look the same because of the control scheme of the FACTS/ESS. But for the PCC bus, the effect with active power output of FACTS/BESS is better than that without active power output.

[FIGURE 17 OMITTED]

[FIGURE 18 OMITTED]

It should be pointed out that the reactive power output of the FACTS/BESS rises when it has exhausted its active power reserve, which is shown in Figure 19. The reason is that it takes more reactive power to improve the voltage of EAF to the same level when no active power is available. In Figure 17, the voltage difference by the two compensation approaches is due to the fact that the 20 MW active power absorbing by ESS causes the voltage drop on the PCC bus.

[FIGURE 19 OMITTED]

Conclusions

The operation of electric arc furnace can cause power quality problems, especially as voltage flickers, to the power supply system to which it is connected. Nowadays, most utilities and power customers are facing the need to solve the power quality problem created by EAF. Some may falsely believe it is only the reactive power demand of the EAF that causes the voltage flicker. This mistake stems from the assumption the X/R ratio is determined mainly by the up- stream transformer or transmission line, which is typically larger than 10. Indeed, for such a high X/R ratio, active power can play a minor role in boosting the bus voltage. The authors have shown the discussion of X/R ratio should be from the point of view of the whole system, which means R and X are not just the impedance of the upper transformer or line, they should be the Thevenin impedance seen from the PCC bus of the whole system. In our study, the active power drawn by EAF also contributes obviously to the voltage flicker. The reason is that actual system X/R ratio is within a range (X/R [approximately equal to] 3 in our sample system) that makes the active power load influencing the voltage drop of the PCC bus. This also proves the need for active power, not only reactive power, to mitigate this kind of voltage problems.

The FACTS with ESS has advantages over FACTS alone by supporting the active and reactive power at the same time. In this paper, the authors analyzed the effects of the FACTS/ESS in mitigation of voltage flickers caused by EAF. A practical EAF model is created to simulate the change of active and reactive powers drawn by an EAF. Different operation modes of the FACTS/ESS have been discussed. The simulation results were presented. The study showed that FACTS with ESS can be more effective than using the FACTS devices alone.

References

[1] Tongxin Zhang, Elham B. Makram, "An Adaptive Arc Furnace Model," IEEE Trans. Power Delivery, Vol. 15, pp. 931-939, July 2000.

[2] Rafael Collantes-Bellido, Tomas Gomez, "Identification and Modeling of a Three Phase Arc Furnace for Voltage Disturbance Simulation," IEEE Trans. Power Delivery, Vol. 12, pp. 1812-1817, October 1997.

[3] Aurelio Garcia-Gerrada, Pablo Garcia-Gonzalez, Rafael Collantes, "Comparison of Thyristor-Controlled Reactors and Voltage-Source Inverters for Compensation of Flicker Caused by Arc Furnaces," IEEE Trans Power Delivery, Vol. 15, pp. 1225-1231, October 2000.

[4] R. Grunbaum, "SVC Light: A Powerful Means for Dynamic Voltage and Power Quality Control in Industry and Distribution", Power Electronics and Variable Speed Drives, pp. 404-409, 18-19 September 2000, Eighth International Conference Publication No.475 @ IEE 2000.

[5] Colin Schauder, "STATCOM for Compensation of Large Electric Arc Furnace Installations", Power Engineering Society Summer Meeting, 1999, IEEE, pp. 1109-1112, July 1999.

[6] Z. Yang, C. Shen, L. Zhang, M. L. Crow and S. Atcitty, "Integration of a StatCom and Battery Energy Storage," IEEE Trans Power Systems, Vol. 16, pp. 254-260, May 2001.

[7] Omer Ozgun and Ali Bur, " Development of an Arc Furnace Model for Power Quality Studies," Power Engineering Society Summer Meeting, 1999 IEEE, pp.507-511, July 1999.

[8] Task Force on Harmonics Modeling and Simulation, Harmonics Working Group, IEEE PES T&D Committee, "Modeling of Devices with Nonlinear Voltage-Current Characteristics for Harmonic Analysis".

[9] Paulo. F. Ribeiro, Aysen Arsoy and Yilu Liu, "Transmission Power Quality Benefits Realized by a SMES-FACTS Controller," Proceedings of 9th International Conference on Harmonics and Quality of Power, pp. 307-312, October 2000.

[10] D. Stade, H. Schau, St. Prinz, "Influence of the Current Control Loops of DC Arc Furnaces on Voltage Fluctuations and Harmonics in the HV Power Supply System", Proceedings of 9th International Conference on Harmonics and Quality of Power, pp. 821-827, October 2000.

Shady A. El-Kashlan * and Hussien El-Desouki Saied

Electrical & Control Engineering Department Arab Academy for Science & Technology (AAST) P.O. Box 1029, Abu-Quir, Alexandria, Egypt * Corresponding e-mail: shosh97@mail.com

Printer friendly Cite/link Email Feedback | |

Author: | Kashlan, Shady A. El-; Saied, Hussien El-Desouki |
---|---|

Publication: | International Journal of Applied Engineering Research |

Article Type: | Report |

Geographic Code: | 1USA |

Date: | Nov 1, 2009 |

Words: | 3610 |

Previous Article: | Intelligent traveling agent system for air lines reservation. |

Next Article: | Comparison of performance and combustion parameters and emissions on a variable compression ratio diesel engine fueled with CsME. |

Topics: |

## Reader Opinion